News for hackers. Stuff that matters.
Re:Inside Pascal: Nvidia 's Newest Computing Platform by semi-extrinsic on Thursday April 21, 2016 @05:01PM (#11545317)
This. Anything with global interactions (i.e. low flops per byte transferred from memory to core) is poorly suited for GPUs.

There is a hierarchy of HPC-type workloads called "Colella's seven dwarves" that ranks different workloads in terms of being CPU bound or memory bandwidth bound. See also the "roofline model". Both of these heuristics are made to reason about CPUs, but are also effective for thinking about GPUs.

There are no comments on this comment yet.